Surgery for liver metastases from primary breast cancer: A reconstructed individual patient data meta-analysis.

ABSTRACT

BACKGROUND: Currently, the outcomes of standard-of-care palliative treatment for BCLM remain poor. Recent literature has shown promising results of hepatic resection, however, not all studies concur. Given the lack of standardized international guidelines in this field, the aim of this study is to provide gold-standard evidence for breast cancer liver metastases (BCLM) through a reconstructed individual patient data meta-analysis approach. METHODS: Four databases were searched for articles comparing surgical and non-surgical treatment for BCLM. One-stage meta-analysis was performed using patient-level survival data reconstructed from Kaplan-Meier curves with plot digitizer software. Shared-frailty and stratified Cox models were fitted to compare survival endpoints. RESULTS: Four propensity-score matched (PSM) studies involving 205 surgical and 291 non-surgical patients for BCLM were included. There was a significant difference between both groups for overall survival (OS) (Hazard Ratio [HR] = 0.40, 95%CI 0.32-0.51). Sensitivity analyses for hormone receptor status of breast cancer (HR = 0.41, 95%CI 0.31-0.55) and type of resection performed (HR = 0.45, 95%CI 0.33-0.61) yielded HRs in favor of surgery. CONCLUSIONS: This meta-analysis concludes that surgery offers superior OS outcomes, compared to non-surgery, in a select group of patients. Future randomized controlled trials and PSM studies are warranted, using this study as a point of reference for similar parameters.